Overview of IR sensor

Based on a simple basic Idea, this proximity sensor, is

easy to build, easy to calibrate and still, it provides a
detection range of 35 cm (range can change depending
on the ambient light intensity).
This sensor can be used for most indoor applications where no
important ambient light is present. For simplicity, this sensor
doesn't provide ambient light immunity, but a more complicated,
ambient light ignoring sensor should be discussed in a coming
article. However, this sensor can be used to measure the speed of
object moving at a very high speed, like in industry or in
tachometers. In such applications, ambient light ignoring sensor,
which rely on sending 40 Khz pulsed signals cannot be used
because there are time gaps between the pulses where the sensor is
'blind'...
The solution proposed doesn't contain any special components, like photo-diodes, photo-transistors, or IR receiver
ICs, only a couple if IR leds, an Op amp, a transistor and a couple of resistors. In need, as the title says, a standard
IR led is used for the purpose of detection. Due to that fact, the circuit is extremely simple, and any novice
electronics hobbyist can easily understand and build it.

Object Detection using IR light

It is the same principle in ALL Infra-Red proximity sensors. The basic idea is to send infra red light through IRLEDs, which is then reflected by any object in front of the sensor.
Then all you have to do is to pick-up the reflected IR light. For detecting the
reflected IR light, we are going to use a very original technique: we are going
to use another IR-LED, to detect the IR light that was emitted from another led of
the exact same type!
This is an electrical property of Light Emitting Diodes (LEDs) which is the fact
that a led Produce a voltage difference across its leads when it is subjected to light.
As if it was a photo-cell, but with much lower output current. In other words, the
voltage generated by the leds can't be - in any way - used to generate electrical
power from light, It can barely be detected. that's why as you will notice in the
schematic, we are going to use a Op-Amp (operational Amplifier) to accurately detect very small voltage changes.

The electronic Circuit

Two different designs are proposed, each one of them is more suitable for different applications. The main
difference between the 2 designs is the way infra-red (IR) light is sent on the object. The receiver part of the
circuit is exactly the same in both designs.
Note: Both the sender and the receiver are constructed on the same board. They are separated in the schematics for
simplification.

Design 1: Low range, Always ON

As the name implies, the sensor is always ON, meaning that the IR led is constantly emitting light. this design of the
circuit is suitable for counting objects, or counting revolutions of a rotating object, that may be of the order of

15,000 rpm or much more. However this design is more power consuming and is not optimized for high ranges. in
this design, range can be from 1 to 10 cm, depending on the ambient light conditions.
As you can see the schematic is divided
into 2 parts the sender and the receiver.
The sender is composed of an IR LED
(D2) in series with a 470 Ohm resistor,
yielding a forward current of 7.5 mA.
The receiver part is more complicated,
the 2 resistors R5 and R6 form a voltage
divider which provides 2.5V at the anode
of the IR LED (here, this led will be used
as a sensor). When IR light falls on the
LED (D1), the voltage drop increases, the
cathode's voltage of D1 may go as low as
1.4V or more, depending on the light
intensity. This voltage drop can be

detected using an Op-Amp (operational Amplifier LM358). You will have to adjust the variable resistor (POT.) R8
so the the voltage at the positive input of the Op-Amp (pin No. 5) would be somewhere near 1.6 Volt. if you
understand the functioning of Op-Amps, you will notice that the output will go High when the volt at the cathode of
D1 drops under 1.6. So the output will be High when IR light is detected, which is the purpose of the receiver.
In case you're not familiar with op-amps, here is shortly and in a very simplified manner, what you need to
know to understand how this sensor functions: The op-amp has 2 input, the +ve input, and the -ve input. If the +ve
input's voltage is higher than the -ve input's voltage, the output goes High (5v, given the supply voltage in the
schematic), otherwise, if the +ve input's voltage is lower than the -ve input's voltage, then the output of the
Op-Amp goes to Low (0V). It doesn't matter how big is the difference between the +ve and -ve inputs, even a
0.0001 volts difference will be detected, and the the output will swing to 0v or 5v according to which input has a
higher voltage.
Some applications of the 'low range Always ON' Design:
Notice how in both devices, the IR leds are encapsulated to protect them from ambient light. this kind of
encapsulation was totally sufficient to overcome all noise due to ambient light for indoor applications.
Wheel Encoder
This is a simple wheel encoder based on the idea that
white stripes will reflect IR light, while black ones will
absorb it. this will result in a series of electrical pulses
as the wheel is rotating, providing the microcontroller
with precious information that can be used to calculate
displacement, velocity or even acceleration. It is now
clear that this kind of sensor has to be Always ON, to
detect every single white stripe passing in front of it, to
achieve accurate results.

Contact-Less tachometer
This is a tachometer, that counts the revolutions per
minute of a rotating object, given that the object has a
reflective stripe glued on it, that will pass in front of
the IR sensor for each and every revolution, giving a
pulse per revolution. Again a microcontroller will have
to be used to 'understand' the data provided by the
sensor and display it. Many commercial contact-less
tachometers, that are sold for more than $200 rely on
this simple idea!
[Build your own one for less than $20 in this article...]

Design 2: High range, Pulsed IR

In this design, which is oriented to obstacle detection in robots, our primary target is to reach high ranges, from 25 to
35 cm, depending on ambient light conditions. The range of the sensor is extended by increasing the current flowing
in the led. This is a delicate task, as we need to send pulses of IR instead of constant IR emission.The duty cycle of
the pulses turning the LED ON and OFF have to be calculated with precision, so that the average current flowing
into the LED never exceeds the LED's maximum DC current (or 10mA as a standard safe value).
The duty cycle is the ratio between the
ON duration of the pulse and the total
period. A low duty cycle will enable us
to inject in the LED high instantaneous
currents while shutting it OFF for
enough time to cool down from the
previous cycle.
Those 2 graphs shows the meaning of
the duty cycle, and the mathematical
relations between the ON time, the Total
period, and the average current.
In the second graph, the average current
in blue is exaggerated to be visible, but
real calculations would yield a much
smaller average current.

Pulsed IR, Duty cycle, Average and Instantaneous current.

Now, hands on the circuit that will put all this theory into practice. The CTRL input in the figure, stands for
Control, and this pin should be connected to the source of the low duty cycle pulses discussed above, whether it is a
microcontroller or an LM555 timer that generates the pulses.
The calculations yielded that a 10 ohm resistor is series with the LED D2, would cause a current of approximately
250 mA to flow through the LED. A current this high, would destroy the LED if applied for a long period of time
(some dozens of seconds), this is why we have to send low duty cycle pulses.
The first Op-amp will provide voltage buffer, to enable any kind of device to control the

sensor, also, it will

provide the 30mA base
current required to drive
the base of the transistor.
The calculation of the the
base resistor R3 depends
on the type of transistor
you use, thus on how
much current you need on
the base to drive the
required collector current.
The receiver part of this
schematic functions in the
exact same way as in the
first design, refer to the
first, 'ALLWAYS ON'
design for a detailed
description.

Software based ambient light detection.

When the sensor is controlled by a microcontroller to generate the low duty cycle pulses, you can benefit from the
High and Low pulses to be able to detect any false readings due to ambient light. This is done by recording 2
different outputs of the sensor, one of them during the ON pulse (the sensor is emitting infra red light) and the other
during the OFF time. and compare the results.
The Idea is enlightened by this graph,
where in the first period, there is low
ambient noise, so the microcontroller
records a "1" during the on cycle,
meaning that an object reflected the
emitted IR Light, and then the
microcontroller records a "0" meaning
that during the OFF time, it didn't receive
anything, which is logic because the
emitter LED was
OFF. But study the second period of the graph, where the sensor is put in a high ambient light environment. As you
can see, the the microcontroller records "1" in both conditions (OFF or ON). This means that we can't be sure
whether the sensor reception was caused by an object that reflected the sent IR light, or it is simply receiving too
much ambient light, and is giving "1" whether there is an obstacle or not.

The following table show the possible outcomes of this method.

Output recorded during:
Software based deduction
On pluse
Off time
1
0
There is definitely an Obstacle in front of the sensor
The sensor is saturated by ambient light, thus we can't know
1
1
if there is an obstacle
There is definitely Nothing in front of the sensor, the way is
0
0
clear
This reading is un logical, there is something wrong with
0
1
the sensor.
Example C Code for 8051 microcontrollers
#include <REGX51.h>
#include <math.h>
unsigned char ir;
bit ir1,ir2;

// to store the final result

// the 2 recording point required for our algorithm

delay(y){

// simple delay function

unsigned int i;
for(i=0;i<y;i++){;}
}
void main(){
//P2.0 IR control pin going to the sensor
//P2.1 IR output pin coming from the sensor
while(1){
P2_0 = 1;
delay(20);
ir1 = P2_1;
P2_0 = 0;
delay(98);
ir2 = P2_1;

}
}

//send IR
//stop IR

if ((ir1 == 1)&(ir2 == 0)){

ir = 1;
// Obstacle detected
P2_3 = 1;
// Pin 3 of PORT 2 will go HIGH turning ON a LED.
if ((ir1 == 1)&(ir2 == 1)){
ir = 2;
// Sensor is saturated by ambient light
}else{
ir = 0;
// The way is clear in front of the sensor.
}

Components positioning:

The correct positioning of the sender LED, the receiver LED with regard to each other and to the Op-Amp can
also increase the performance of the sensor. First, we need to adjust the position of the sender LED with respect to
the receiver LED, in such a way they are as near as possible to each others , while preventing any IR light to be
picked up by the receiver LED before it hit and object and returns back. The easiest way to do that is to put the
sender(s) LED(s) from one side of the PCB, and the receiver LED from the other side, as shown in the 3D model
below.
This 3D model shows the position of the LEDs. The
green plate is the PCB holding the electronic components
of the sensor. you can notice that the receiver LED is
positioned under the PCB, this way, there wont be ambient
light falling directly on it, as ambient light usually comes
from the top.
It is also clear that this way of positioning the LEDs
prevent the emitted IR light to be detected before hitting an
eventual obstacle.
Another important issue about components positioning, is the distance between the receiver LED and the Op-Amp.
which should be as small as possible. Generally speaking, the length of wires or PCB tracks before an amplifier
should be reduced, otherwise, the amplifier will amplify - along with the original signal - a lot of noise picked up
form the electromagnetic waves traveling the surrounding.
Here is an example PCB where the distance between the LED and the OpAmp is shown. Sure this distance is not as critical as you may think, it can be
up to 35mm without causing serious problems, but trying to reduce this
distance will Always give you better results.
Actually, when I design the PCB, I start by placing the receiver LED and the
Op-Amp, as near to each others as possible, then continue the rest of the
design.

An example PCB construction

Here is an example construction of the PCB for the High Range, Pulsed IR proximity sensor.
You can download here the project folder containing the schematic, the PCB design, and an example code for 8051
microcontroller to send the low duty cycle pulses.

In this design, the LM358 Op-Amp is

mounded on the copper side, to save
some space. The POT is the
potentiometer used to adjust sensitivity.
As explained before, the sender and
receiver LEDs are on both sides of the
PCB.

Testing the High range Pulsed IR sensor

The last step, is to test the performance of the pulsed IR proximity sensor. To do this, I connected the sensor to a
89S52 microcontroller, loaded with a program to generate pulses with a duty cycle of approximately 1.6. at a
frequency of 3Khz. LEDs are deigned to operate at very high frequencies, so you don't have to worry about the
response time. To make sure your duty cycle calculations are correct, let the sensor running for a minute, and check
with your fingers the temperature of the IR sender LED. If its not hot, then everything is alright. On the other hand,
if the LED is getting hot, to an extent that you can feel it, there is probably something wrong, you should then try to
decrease the duty cycle, or increase the series resistor, in order to decrease the average current flowing into the LED.
Then, you can start testing the range of the sensor, and experiment it in different ambient light conditions, but the
potentiometer may have to be adjusted carefully, to cope with ambient light.

In the example C code above, the final output of the sensor appears on the pin P2_3 of the microcontroller